Purpose: We examined whether early response/non-response to risperidone according to the Clinical Global Impressions-improvement scale (CGI-I) at 2 weeks could predict subsequent response. This prediction was also applied to olanzapine. We then investigated whether early non-responders (ENRs) to risperidone or olanzapine who switched to the other showed significantly greater improvement, compared with those staying on the initial antipsychotic. We performed a rater-blinded, randomized controlled trial in 18 psychiatric emergency sites. Eligible patients were newly admitted patients with acute schizophrenia. Early response was defined as CGI-I ≤ 3 following 2 weeks of treatment. The primary outcome measure was achievement of remission and ≥ 50% improvement in the Positive and Negative Syndrome Scale at 4 weeks.
Results: At 4 weeks, 53% of risperidone early responders (ERs) went into remission, whereas only 9% of ENRs staying on risperidone (n=11) did (P=0.016). Similarly, at 4 weeks, 81% of risperidone ERs achieved ≥ 50% response, whereas only 9% of ENRs staying on risperidone achieved ≥ 50% response (P < 0.0001). In contrast, 58% of olanzapine ERs (n=33) went into remission, whereas 38% of ENRs staying on olanzapine (n=8) did at 4 weeks (P=0.44). Similarly, 61% of olanzapine ERs achieved ≥ 50% response, whereas 25% of ENRs staying on olanzapine achieved ≥ 50% response (P=0.12). The negative likelihood ratio for the prediction of ≥ 50% response at 4 weeks by early response status to risperidone at 2 weeks was 0.057.
Conclusion: In newly admitted patients with acute schizophrenia, non-response to risperidone using CGI-I at 2 weeks can predict subsequent response. It looks like there is significant response to olanzapine that doesn't occur until 4 weeks. Thus, clinicians may want to switch to another drug earlier when risperidone is the first drug, and later when olanzapine is the first drug.
Copyright © 2011 Elsevier B.V. All rights reserved.